The generation of SARS-CoV-2 virus-like particles in insect cells

By Namita MitraMay 2 2022Reviewed by Danielle Ellis, B.Sc.

The rapid spread of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), resulting in the COVID-19 pandemic, has pushed the world towards developing rushed but effective vaccines against the disease. Most currently employed COVID-19 vaccines rely on the principle of virus inactivation, adenoviral vectors, or the use of direct nucleic acids (mRNA/DNA).

In a new study recently published in the journal Viruses, researchers have reported the development of a promising COVID-19 vaccine candidate which is based on SARS-CoV-2 virus-like particles (VLPs), derived from the co-expression of the spike (S), membrane (M) and envelope (E) structural proteins in the baculovirus expression system.

In vitro testing confirmed their antigenicity, and viral challenge in VLP-immunized hamsters demonstrated their immunogenicity. Immunization with the SARS‐CoV‐2 VLPs could not prevent in vivo replication of the challenge virus; however, the viral titers and disease pathology markers were reduced compared to the control group.

The virus-like particles (VLPs)

VLPs correspond to structures that are similar to the virus; however, they are replication-deficient because of the absence of genomic DNA or RNA. A VLP can also be imagined as an empty virus shell. These can directly interact with antigen-presenting cells (APCs), especially the dendritic cells, which are the most potent APCs.

Coronaviral M and E proteins, if co‐expressed in vitro, can result in the formation of VLPs. The membrane-spanning protein-M, the most abundant of the coronavirus-envelope associated proteins, primarily drives the assembly of coronavirus particles. While M oligomerizes to form a lattice that structures the virus envelope's conformation, it also recruits other structural proteins into the nascent viral particles completing the assembly process. The smaller envelope E protein, though incorporated into the nascent virions in very minute quantity, is indispensable for successful virus budding.

The spike S trimeric protein, responsible for interaction with the host receptor ACE2 and facilitating membrane fusion, can also be co‐expressed in the cells expressing M and E. S is a highly immunogenic antigen that drives the generation of neutralizing antibodies, blocking the interaction between S1 and ACE2.

As the VLPs mimic the authentic virus structure and express trimeric S protein at multiple sites, they appropriately stimulate the humoral immunity and are amenable to being recognized by the antigen-presenting cells (APCs) to generate a suitable T‐cell response. Also, co-administration of an adjuvant is not required due to their multimeric nature.

Several VLP based vaccines for other viruses, produced by the baculovirus insect cell system, are currently in use, proving evidence of scalability and acceptability", the team highlights.

SARS-CoV-2 VLP construction and verification

The team, in the past, had constructed a recombinant baculovirus that expressed S, E, and M structural proteins required for SARS‐CoV VLP formation. Following a similar strategy, the team in the current study developed a single recombinant baculovirus expressing SARS-CoV-2 S, M and previously used SARS-CoV E protein. The E proteins of SARS‐CoV and SARS‐CoV‐2 are functionally exchangeable for the purposes of VLP production. Therefore, the team used the existing SARS-CoV E gene insert for the new recombinant virus construction.

Western blotting of insect cells infected with the recombinant virus (VLP competent construction) confirmed the expression of both S and M proteins together. Also, the S protein was demonstrated to exist as a trimeric protein (the native form on viral envelope) in the VLP structure. VLPs were purified on sucrose gradient where they form a distinct band at 35% sucrose gradient. Coomassie blue-stained SDS‐PAGE on the 35% gradient fraction showed 3 prominent bands at 10, 30 and 180 kDa corresponding to E, M and S proteins, respectively.

Therefore, the team successfully produced a single recombinant baculovirus that could express, all by itself, the three structural proteins required for SARS‐CoV‐2 VLP formation.

Transmission electron microscopy (TEM) of the fraction showed VLPs assembled as vesicle-like structures of ~100 nm diameter displaying crown‐like spikes characteristic of coronavirus particles.

SARS-CoV-2 VLPs are antigenic as well as immunogenic

Convalescent sera were used to probe the VLPs for antigenicity. Multiple convalescent sera reacted strongly with the immobilized VLPs on ELISA, suggesting that VLPs displayed S antigen in a form suitable to bind the antibodies.

A syrian hamster model, which displays a disease pathology and neutralizing antibody response similar to human COVID-19, was used to test the Immunogenicity of the VLPs bearing the SARS‐CoV‐ 2 S protein. VLP candidate vaccination was administered in two doses 28 days apart to 5 hamsters. No treatment controls were also kept.

Seroconversion and the development of neutralizing antibodies to RBD were observed in 4 out of 5 animals after the first dose and in all VLP immunized animals after the second dose. The control group did not show any antibody activity.

Later, two weeks after the second VLP dose, the hamsters were experimentally infected with an infectious dose of the SARS‐CoV‐2 and observed the antibody activity. Neutralizing antibody titers elevated within 4 days of challenge in the VLP immunized animals earlier than the control group.

The team also observed a slower clearance of genomic RNA from the oral swabs of controls compared to the VLP immunized group. The peak virus load on challenge day 2 in oral swabs of the treatment group was much lower than that in the control group, i.e., 1.4 × 10⁷ and 4.7 × 10^7, respectively. However, the difference was not significant. Nasal swabs also showed a similar pattern.

At day 10, the lungs of VLP immunized animals showed lower scores for the markers of inflammation and lesser S positive syncytial cells than the non-vaccinated controls.

All treatment and control animals recovered by day 14 post virus challenge but the immunized animals recovered significantly faster than the controls.

The findings suggested that antibody levels were insufficient to protect from infection but were sufficient to reduce disease severity.

In our study, VLPs were immunogenic in the absence of adjuvant and were produced using a technology that is already in use for both human and animal vaccines offering scalability, an acceptable manufacturing process and an established route to licensure", concludes the team.